CN115469260B - Hausdorff-based current transformer anomaly identification method and system - Google Patents
Hausdorff-based current transformer anomaly identification method and system Download PDFInfo
- Publication number
- CN115469260B CN115469260B CN202211203322.7A CN202211203322A CN115469260B CN 115469260 B CN115469260 B CN 115469260B CN 202211203322 A CN202211203322 A CN 202211203322A CN 115469260 B CN115469260 B CN 115469260B
- Authority
- CN
- China
- Prior art keywords
- current
- line
- phase
- hausdorff
- constructing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 33
- 238000011156 evaluation Methods 0.000 claims abstract description 32
- 238000000819 phase cycle Methods 0.000 claims abstract description 23
- 230000005856 abnormality Effects 0.000 claims abstract description 15
- 238000003745 diagnosis Methods 0.000 claims abstract description 13
- 238000012216 screening Methods 0.000 claims abstract description 12
- 230000002159 abnormal effect Effects 0.000 claims description 28
- 238000004364 calculation method Methods 0.000 claims description 25
- 239000013598 vector Substances 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 12
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 12
- 239000010931 gold Substances 0.000 claims description 12
- 229910052737 gold Inorganic materials 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 11
- 238000012847 principal component analysis method Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 4
- 239000000463 material Substances 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 16
- 239000011159 matrix material Substances 0.000 description 14
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000005259 measurement Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000000513 principal component analysis Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000004804 winding Methods 0.000 description 2
- 241000287196 Asthenes Species 0.000 description 1
- 125000003275 alpha amino acid group Chemical group 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/02—Testing or calibrating of apparatus covered by the other groups of this subclass of auxiliary devices, e.g. of instrument transformers according to prescribed transformation ratio, phase angle, or wattage rating
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/50—Arrangements for eliminating or reducing asymmetry in polyphase networks
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a Hausdorff-based current transformer anomaly identification method, which comprises the following steps: acquiring running current data of a current transformer, carrying out multi-order difference on the running current data, and screening stable current data; constructing three-phase asymmetric current components of the lines based on the stable current data, and calculating an evaluation statistic and variation of each line; constructing a line anomaly identification model and a phase sequence diagnosis model through different SVM algorithms according to the Hausdorff distance of the unbalance degree of the three-phase asymmetric current components of the line, the variation of the evaluation statistic and the kirchhoff current law; and carrying out anomaly identification on the current transformer by using the line anomaly identification model and the phase sequence diagnosis model. The invention combines Hausdorff distance and PCA to extract current characteristics, and identifies the abnormality of the current transformer through a plurality of SVM models, thereby realizing the on-line monitoring of the metering error state of the current transformer.
Description
Technical Field
The invention belongs to the technical field of power equipment measurement, and particularly relates to a Hausdorff-based current transformer anomaly identification method and system.
Background
The current transformer (Current transformers) is an important measurement device in the power system. The primary winding is connected in series in the power transmission and transformation main loop, and the secondary winding is connected into equipment such as a measuring instrument, a relay protection device or an automatic device according to different requirements, and the like, and is used for converting the large current of the primary loop into the small current of the secondary side so as to be used for the safety acquisition of measurement and control protection metering equipment. The method is accurate and reliable, and has great significance for safe operation, control protection, electric energy metering and trade settlement of the electric power system.
Different from a voltage transformer, a current transformer is characterized in that: 1. the physical relationship in the current transformer group is relatively complex, the physical constraint condition is hidden, and the physical relationship of the voltage transformer group is highlighted. The measurement values of the voltage transformers with the same phase at the same node in the transformer substation are consistent, the mutual comparison measurement values among groups can be used for judging, and the line currents are independent and cannot be realized by mutual comparison of the measurement values. 2. The voltage amplitude in the steady-state transformer substation is 110% -120% of rated voltage change, the voltage fluctuation is small, the overall data characteristics of the voltages in different transformer substations are consistent, the voltage information characteristics are obvious and universal, and the information physical fusion based on the voltage signals is easy to realize. The circuit current changes in different substations are mutually independent, the amplitude is changed from 0% -120% of rated current, the fluctuation is extremely large, and the circuit current has information characteristics and is hidden, so that the on-line monitoring of the metering error state of the current transformer is difficult to realize.
Disclosure of Invention
In order to solve the problem of difficult online monitoring of the metering error state of the current transformer, the first aspect of the invention provides a Hausdorff-based current transformer anomaly identification method, which comprises the following steps: acquiring running current data of a current transformer, carrying out multi-order difference on the running current data, and screening stable current data; constructing three-phase asymmetric current components of one or more lines based on the stationary current data; according to the three-phase asymmetric current components, calculating Hausdorff distance of unbalance degree of each line and other three-phase asymmetric current components of the same bus; constructing characteristic parameters of line current based on the stable current data, and constructing a calculation model by using a principal component analysis method; calculating the evaluation statistic and the variation of each line according to the calculation model; constructing a line anomaly identification model by using a first SVM algorithm according to Hausdorff distance of unbalance degree of three-phase asymmetric current components of each line and other lines under the same bus and the variation of evaluation statistics; constructing a phase sequence diagnosis model by using a second SVM algorithm based on a kirchhoff current law and the Hausdorff distance of an abnormal line; and carrying out abnormality recognition on one or more current transformers of the line to be evaluated by using the line abnormality recognition model and the phase sequence diagnosis model.
In some embodiments of the present invention, the constructing a line anomaly identification model by using a first SVM algorithm according to Hausdorff distance of unbalance degree of three-phase asymmetric current components of each line and other lines under the same bus and variation amount of evaluation statistic includes: constructing a feature vector according to the Hausdorff distance of the unbalance degree of the three-phase asymmetric current components of each line and other lines under the same bus and the variation of the evaluation statistic; determining an objective function and a kernel function of a first SVM algorithm, and constructing a line anomaly identification model according to the objective function and the kernel function; the objective function is expressed as:
，
wherein, As weights, y j epsilon { +1, -1} represents a category label of which the line j is normal and abnormal, and v j is a feature vector of the line j; is a threshold value; c represents a penalty coefficient and, Represents a relaxation factor; the kernel function is expressed as:， Is a gaussian radial basis function.
Further, a gold eagle optimization algorithm is adopted to optimize C, g parameters in the first SVM model.
In some embodiments of the present invention, the constructing the phase sequence diagnostic model using the kirchhoff current law and Hausdorff distance of the abnormal line includes: based on kirchhoff current law, calculating Hausdorff distance between three-phase current of an abnormal line and three-phase current of other lines on the same node; calculating the contribution rate variation of each phase of current in the abnormal line to the evaluation statistic; and constructing a phase sequence recognition model by adopting a second SVM based on the contribution rate variation and the Hausdorff distance.
Further, the contribution rate is calculated by:
，
wherein, The first contribution rate array cont (t) at time tA respective element, which is also the firstThe contribution rate of the station current transformer to the statistic Q (t); Denoted as time t Real-time data standardized by the mutual inductor; Is that Projection in principal component space.
In the foregoing embodiment, calculating, according to the three-phase asymmetric current components, a Hausdorff distance of an imbalance degree of the three-phase asymmetric current components of each line and other lines under the same bus includes: calculating zero sequence unbalance and negative sequence unbalance of each line based on the three-phase asymmetric current components; and calculating Hausdorff distance between the zero sequence imbalance characteristic parameters of each line and other lines in the same period for the lines on the same bus.
In a second aspect of the present invention, there is provided a Hausdorff-based current transformer anomaly identification system, comprising: the acquisition module is used for acquiring the running current data of the current transformer, carrying out multi-order difference on the running current data and screening out stable current data; constructing three-phase asymmetric current components of one or more lines based on the stationary current data; according to the three-phase asymmetric current components, calculating Hausdorff distance of unbalance degree of each line and other three-phase asymmetric current components of the same bus; the calculation module is used for constructing characteristic parameters of line current based on the stable current data and constructing a calculation model by utilizing a principal component analysis method; calculating the evaluation statistic and the variation of each line according to the calculation model; the first construction module is used for constructing a line anomaly identification model by utilizing a first SVM algorithm according to the Hausdorff distance of the unbalance degree of the three-phase asymmetric current components of each line and other lines under the same bus and the variation of the evaluation statistic; the second construction module is used for constructing a phase sequence diagnosis model by using a second SVM algorithm based on the kirchhoff current law and the Hausdorff distance of the abnormal line; and the identification module is used for carrying out abnormality identification on one or more current transformers of the line to be evaluated by utilizing the line abnormality identification model and the phase sequence diagnosis model.
In a third aspect of the present invention, there is provided an electronic apparatus comprising: one or more processors; and the storage device is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors are enabled to realize the Hausdorff-based current transformer anomaly identification method provided by the first aspect of the invention.
In a fourth aspect of the present invention, there is provided a computer readable medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the Hausdorff-based current transformer anomaly identification method provided in the first aspect of the present invention.
The beneficial effects of the invention are as follows:
The invention provides a Hausdorff-based current transformer anomaly identification method, which comprises the steps of screening stabilized segment data according to the measuring range and current fluctuation of a current transformer, respectively adopting a Hausdorff distance algorithm and a Q statistic algorithm to calculate the Hausdorff distance ratio and the Q statistic variation delta Q of a distance from a line according to preprocessed stabilized three-phase current data; then an improved SVM algorithm is adopted to construct a line anomaly identification model; based on an abnormal line model, a Hausdorff distance algorithm and a contribution rate index variable delta cont i (t) are adopted to construct an abnormal phase sequence identification model, so that the on-line identification of the abnormal transformer from line current data with large fluctuation and hidden characteristics is realized.
Drawings
FIG. 1 is a basic flow diagram of a Hausdorff-based current transformer anomaly identification method in some embodiments of the invention;
FIG. 2 is a schematic diagram of a specific flow chart of a Hausdorff-based current transformer anomaly identification method in some embodiments of the invention;
FIG. 3 is a schematic diagram of a Hausdorff-based current transformer anomaly identification system in some embodiments of the invention;
Fig. 4 is a schematic structural diagram of an electronic device according to some embodiments of the present invention.
Detailed Description
The principles and features of the present invention are described below with reference to the drawings, the examples are illustrated for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.
Referring to fig. 1, in a first aspect of the present invention, there is provided a Hausdorff-based current transformer anomaly identification method, including: s100, acquiring running current data of a current transformer, carrying out multi-order difference on the running current data, and screening out stable current data; constructing three-phase asymmetric current components of one or more lines based on the stationary current data; according to the three-phase asymmetric current components, calculating Hausdorff distance of unbalance degree of each line and other three-phase asymmetric current components of the same bus; s200, constructing characteristic parameters of line current based on the stable current data, and constructing a calculation model by using a principal component analysis method; calculating the evaluation statistic and the variation of each line according to the calculation model; s300, constructing a line anomaly identification model by using a first SVM algorithm according to Hausdorff distance of unbalance degree of three-phase asymmetric current components of each line and other lines under the same bus and the variation of evaluation statistics; s400, constructing a phase sequence diagnosis model by using a second SVM algorithm based on a kirchhoff current law and a Hausdorff distance of an abnormal line; s500, carrying out abnormality recognition on one or more current transformers of the line to be evaluated by using a line abnormality recognition model and a phase sequence diagnosis model.
In step S100 of some embodiments of the present invention, operation current data of a current transformer is obtained, and multi-order difference is performed on the operation current data and stationary current data is screened out; specifically, collecting running current data of the current transformer, preprocessing the current data by adopting first-order difference and second-order difference, and screening stable current data;
For the current transformer, when the line current is lower than the rated current, the error of the current transformer is larger, and the data quality is lower, so that current data with the rated range of 80% -120% is screened. Meanwhile, the current fluctuation in the power grid is large, and more data breakpoints exist in the current data, so that the collected current amplitude data is subjected to first-order and second-order differential processing according to the formula (1) and the formula (2), and the current data breakpoints are screened out.
First order difference:； （1），
Second order difference: ； （2），
Wherein x (omega) is current amplitude data, delta 1x(Ω)、Δ2 x (omega) is a first-order and second-order differential value of the current amplitude data, and omega is a data point. When (when) 、When the first-order data and the second-order data are stable, the first-order data and the second-order data are judged to be stable、Is a set threshold.
It can be appreciated that the smoothed current data is screened out based on the first and second order differential results. Constructing three-phase asymmetric current components of one or more lines based on the stationary current data; according to the three-phase asymmetric current components, calculating Hausdorff distance of unbalance degree of each line and other three-phase asymmetric current components of the same bus; specifically, based on the screened current data, constructing a zero sequence current component and a negative sequence current component of the circuit, and calculating the ratio between the circuits by respectively adopting Hausdorff distance、; The current data in the power grid has large fluctuation, transient processes in the power grid frequently occur, amplitude fluctuation is large, no fixed rule exists, and the construction of an operation error monitoring model is difficult to realize according to amplitude phase characteristics. However, the negative sequence unbalance and the zero sequence unbalance are relatively stable, and a certain rule exists, so that model construction can be realized according to the negative sequence unbalance and the zero sequence unbalance as characteristic quantities.
Specifically, according to the pre-screened three-phase current modeling data, the asymmetric three-phase current phasors are decomposed into symmetric positive and negative sequence and zero sequence current components according to a formula (3).
（3）,
Wherein I a is to select a phase as a reference phase,Namely, the three-phase current is obtained,Is the corresponding positive sequence, negative sequence and zero sequence component of a phase. In the operation operator,. And:
Then: 。
then, the characteristic parameters are extracted: zero sequence imbalance and negative sequence imbalance;
Zero sequence imbalance:
（4），
negative sequence imbalance:
（5），
And obtaining the zero sequence unbalance degree and the negative sequence unbalance degree of the circuit based on the formula (4) and the formula (5).
2) Hausdorff distance algorithm principle
Hausdorff distance is a measure that describes the degree of similarity between 2 sets of points. Assume that there are 2 sets of points:
；Z={z1,z2,z3,…,zq} (6)；
the Hausdorff distance between U, Z is then:
H(U,Z)=max(h(U,Z),h(Z,U)) （7），
wherein, 。
3) And calculating Hausdorff distance between the zero sequence imbalance and negative sequence imbalance characteristic parameters of each line in the same period for the lines on the same bus.
① Taking zero sequence unbalance as an example, calculating Hausdorff distance between the zero sequence unbalance of each line:
（8），
in the formula (8), n represents the number of lines on the same bus, and H n1 is the Hausdorff distance between the zero sequence unbalance degree of the nth line and the 1 st line.
Based on the matrix H, calculating a j-th Hausdorff distance ratio: calculating the ratio:（9），
In formula (9), r j is the ratio of the j-th column, H maxj=max{H1j,H2j …Hnj, m=1, 2, …, n; j=1, 2, …, n, n is the number of lines; j+.n, r j is the ratio of the j-th column.
② Calculating the ratio of Hausdorff distance between the zero sequence unbalance degree and the negative sequence unbalance degree of each line based on the step ①、。
In step S200 of some embodiments of the present invention, a characteristic parameter of the line current is constructed based on the stationary current data, and a calculation model is constructed using a principal component analysis method; calculating the evaluation statistic and the variation of each line according to the calculation model; specifically, based on screening stable current data, line current is used as a characteristic parameter, a PCA (principal component analysis) is adopted to construct a calculation model, and an evaluation standard quantity under a normal mode is calculatedThe line real-time statistic Q (t) is used for calculating the variation delta Q (t) of each line statistic; the more detailed steps are as follows:
Taking current data at 80% -120% of rated amplitude as characteristic parameters, constructing an evaluation calculation model under a normal mode by adopting PCA, and calculating evaluation statistics 。
1) Constructing a sample set under a normal mode by using current data under the normal mode
（10），
Where N is the number of sampling points. A. B, C denotes three phases, and x denotes sample current data in a sample.
2) Normalization process
To avoid the influence caused by the difference of variable dimensions, the obtained data sample is needed to be obtainedAnd (3) carrying out standardization processing, wherein the standardized data matrix is as follows:
（11），
wherein N is the number of sampling points, and M is the number of transformers. ， WhereinIs the mean of the M-th column vector of matrix Y 0,WhereinIs the variance of the M-th column vector of matrix Y 0.
3) Based onSingular value decomposition is performed on covariance R of (c) to determine a load matrix p e of the residual subspace. From a modeling dataset in a normal modalityAnd the load matrix p e of the corresponding residual subspace calculates the evaluation standard quantity under the confidence coefficient by adopting a method based on kernel density estimation; From real-time data setsAnd the corresponding residual subspace payload matrix p e computes the real-time statistic Q (t).
（12），
Wherein, the left side R is covariance matrix, the right side is singular value decomposition,Is the eigenvalue of covariance matrix, and the arrangement order satisfiesP m pe represents a feature vector matrix; the eigenvector matrix [ P m pe ] obtained at this time is the load matrix P. Load matrix p m of principal component is formed by accumulated variance contribution rate, and the rest form residual load matrix。
4) The Q statistic is embodied as follows:
（13），
calculating an evaluation statistic of the line based on the equation (11) and the equation (12) . And acquiring real-time running data of the lines, and calculating real-time statistics Q (t) of each line based on the model.
5) Based on evaluation statisticsAnd calculating the statistic variation delta Q (t) of each line according to the real-time statistic Q (t).
In step S300 of some embodiments of the present invention, the constructing a line anomaly identification model by using the first SVM algorithm includes: s301, constructing a feature vector according to Hausdorff distance of unbalance degree of three-phase asymmetric current components of each line and other lines under the same bus and the variation delta Q (t) of evaluation statistics;
Based on 、And constructing a feature vector by delta Q (t), and constructing an identification model by adopting an SVM algorithm. To be used for、And constructing a characteristic vector v at the moment t by delta Q (t):
（14）；
in the formula (14), the amino acid sequence of the compound, Represents the zero sequence unbalanced Hausdorff distance ratio of the jth line to other lines,The negative sequence unbalance Hausdorff distance ratio of the jth line to other lines is represented, and DeltaQ (t) is the variation of the statistic of the jth line; j=1, 2, …, n, n is the number of lines.
Based on characteristic parameters of the samples, an SVM model is adopted to search an optimal hyperplane capable of completely separating samples of different categories. Considering that outliers in the data can seriously affect the classification performance of the SVM, in order to make the model more robust, soft interval and penalty terms are introduced, and the improved SVM objective function is as follows:
（15），
wherein, As weights, y j epsilon { +1, -1} represents a category label of which the line j is normal and abnormal, and v j is a feature vector of the line j; is a threshold value; c represents a penalty coefficient and, Represents a relaxation factor; another important part of the SVM is a kernel function and its kernel function parameters, which can help the SVM handle the non-linearity problem that it cannot solve. The classification function of the SVM in the case of kernel mapping is:
（16），
wherein, Is a Lagrange multiplier, and the Lagrange multiplier,As a gaussian radial basis function,. The key of the SVM model construction is to solve the optimal value problem of the core parameter g and the penalty coefficient C. Therefore, the model performance is improved by introducing a gold eagle optimization algorithm to optimize SVM model parameters. The method comprises the following specific steps:
1) Attack behavior
The attack vectors of gold hawk are:
（17），
In the method, in the process of the invention, Is the firstOnly the attack vector of gold hawk,For the best hunting site (prey) reached by the current gold eagle,Is the firstOnly gold eagle is currently located.
2) Cruising behaviour
The scalar form of the hyperplane in three dimensions is:
（18），
In the method, in the process of the invention, As a normal vector of the sample,Is a variable vector. Here, theRepresenting the position of the gold eagle, s 1 is defined as the penalty factor C in the SVM model, and s 2 is the kernel parameter g in the SVM model. Searching for the value of the fixed variable:
（19），
Where c k is the kth element of the target point, Is the first of attack vectorsThe element, a k, is the kth vector of the attack vector. Random target points on the flight hyperplane can be found accordingly. The general representation of the target point is:
（20），
Wherein, random E [0,1], the random number is updated to enable gold hawk to be explored to a random target point.
3) Move to a new position
The displacement of the gold hawk consists of an attack vector and a target position, and the iteration step size vector is as follows:
（21），
Where p a is the attack coefficient, p c is the cruise coefficient, 、Is a random vector within [0,1 ].
Based on this the next position of the gold eagle can be found:
（22），
Is the first Only the t+1 th time position of the gold hawk,Is the firstThe position is Jin Yingdi t times only,Is the firstOnly Jin Yingdi t moves the step size. Then, the updated formulas of the attack coefficient p a and the cruise coefficient p c are:
（23），
Where T represents the current iteration number and T represents the maximum iteration number. 、The initial and final values of p a respectively,、The initial and final values of p c, respectively. In order to prevent the optimization algorithm from falling into local optimum, the eagle position updating algorithm is optimized:
（24），
is a random vector in [0,1], rand is a random factor obeying uniform distribution on [0,1], Is a fixed parameter. And comparing the fitness values of the two strategies, and selecting the strategy with the better fitness as the eagle position updating strategy.
In step S400 of some embodiments of the present invention, constructing a phase sequence diagnostic model using the kirchhoff current law and the Hausdorff distance of the abnormal line by using the second SVM algorithm includes: based on kirchhoff current law, calculating Hausdorff distance between three-phase current of an abnormal line and three-phase current of other lines on the same node; calculating the contribution rate variation of each phase of current in the abnormal line to the evaluation statistic; and constructing a phase sequence recognition model by adopting a second SVM based on the contribution rate variation and the Hausdorff distance.
Specifically, based on kirchhoff's current law, hausdorff distances H ψ between three phases of an abnormal line A, B, C and three phases of the rest line A, B, C on the same node are calculated respectively; defining a line L 1、L2…Ll as a power transmission line of each branch connected with a bus, wherein the three phases A, B, C of the line all meet the kirchhoff theorem:
（25），
Wherein the method comprises the steps of Representing a sequence of phase a primary current sample values for line L 1. Based on the above line locating result, selecting an abnormal line L y and recording the A-phase primary current vector of the L y line as. The sum of the phasors of the primary currents of the remaining lines is recorded asThen
（26），
I.e.And (3) withEqual amplitude and phase differenceThen based on Hausdorff distance algorithm:
（27），
wherein, 、Respectively represent pairs ofAnd (3) withThe current sampling sequence performs per unit current data.
In the actual operation process, the influence of the operation error of the transformer is considered, and the calculation is performed through the secondary side data of the current transformer:
（28），
wherein, Secondary current data per unit of current data is collected for the a phase of the selected anomaly line L y,A nominal transformation ratio for line L y; And calculating current phasors and per-unit current data for other lines of the same bus through secondary side current and rated transformation ratio. The distances of the three phases (one or more phases) of the abnormal line A, B, C from the in-phase current data of the rest lines during a period of operation time are calculated.
（29），
Further, the contribution rate is calculated by:
，(30)
wherein, The first contribution rate array cont (t) at time tA respective element, which is also the firstThe contribution rate of the station current transformer to the statistic Q (t); Denoted as time t Real-time data standardized by the mutual inductor; Is that Projection in principal component space.
In the foregoing embodiment, calculating, according to the three-phase asymmetric current components, a Hausdorff distance of an imbalance degree of the three-phase asymmetric current components of each line and other lines under the same bus includes: calculating zero sequence unbalance and negative sequence unbalance of each line based on the three-phase asymmetric current components; and calculating Hausdorff distance between the zero sequence imbalance characteristic parameters of each line and other lines in the same period for the lines on the same bus.
Referring to fig. 2, in one embodiment of the present invention, the method for identifying abnormality of a current transformer based on Hausdorff includes: step A: collecting running current data of the current transformer, preprocessing the current data by adopting first-order difference and second-order difference, and screening stable current data;
step A: based on the screened current data, constructing a zero sequence current component and a negative sequence current component of the circuit, and calculating the ratio between the circuits by adopting Hausdorff distance 、；
And (B) step (B): based on screening stable current data, constructing a calculation model by taking line current as a characteristic parameter and adopting PCA (principal component analysis), calculating an evaluation standard quantity Q α and line real-time statistic Q (t) under a normal mode, and calculating the variation delta Q (t) of each line statistic;
step C: calculating each line based on step B, C 、And delta Q (t), and constructing a line anomaly identification model by adopting an improved GEO-SVM learning algorithm.
Step D: based on kirchhoff current law, hausdorff distances between three phases of an abnormal line A, B, C and three phases of the rest line A, B, C on the same node are calculated respectively；
Step E: calculating A, B, C three-phase relative statistics in abnormal lineAmount of change in contribution rate of (2)；
Step F: based on the contribution rate variationAnd Hausdorff distance H i, constructing a phase sequence identification model by adopting an SVM, and positioning the phase sequence number of the abnormal transformer in the abnormal line;
step G: inputting the line to be evaluated into the model to realize the abnormal identification of the transformer.
Example 2
Referring to fig. 3, in a second aspect of the present invention, there is provided a Hausdorff-based current transformer anomaly identification system 1, comprising: the acquisition module 11 is used for acquiring the running current data of the current transformer, carrying out multi-order difference on the running current data and screening out stable current data; constructing three-phase asymmetric current components of one or more lines based on the stationary current data; according to the three-phase asymmetric current components, calculating Hausdorff distance of unbalance degree of each line and other three-phase asymmetric current components of the same bus; a calculation module 12, configured to construct a characteristic parameter of the line current based on the stationary current data, and construct a calculation model using a principal component analysis method; calculating the evaluation statistic and the variation of each line according to the calculation model; the first construction module 13 is configured to construct a line anomaly identification model by using a first SVM algorithm according to Hausdorff distance of unbalance degree of three-phase asymmetric current components of each line and other lines under the same bus and variation amount of evaluation statistic; a second construction module 14, configured to construct a phase sequence diagnostic model based on kirchhoff's current law and Hausdorff's distance of the abnormal line by using a second SVM algorithm; the identifying module 15 is configured to identify an abnormality of one or more current transformers of the line to be evaluated by using the line abnormality identifying model and the phase sequence diagnosing model.
Further, the second building block 14 includes: the first calculation unit is used for calculating Hausdorff distance between the three-phase current of the abnormal line and the three-phase current of the other lines on the same node based on the kirchhoff current law; a second calculation unit for calculating a contribution rate variation of each phase of current in the abnormal line to the evaluation statistic; and the construction unit is used for constructing a phase sequence identification model by adopting a second SVM based on the contribution rate variation and the Hausdorff distance.
Example 3
Referring to fig. 4, a third aspect of the present invention provides an electronic device, including: one or more processors; and storage means for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the method of the present invention in the first aspect.
The electronic device 500 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 501 that may perform various appropriate actions and processes in accordance with programs stored in a Read Only Memory (ROM) 502 or loaded from a storage 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic apparatus 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.
The following devices may be connected to the I/O interface 505 in general: input devices 506 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 507 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 508 including, for example, a hard disk; and communication means 509. The communication means 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. While fig. 4 shows an electronic device 500 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead. Each block shown in fig. 4 may represent one device or a plurality of devices as needed.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 509, or from the storage means 508, or from the ROM 502. The above-described functions defined in the methods of the embodiments of the present disclosure are performed when the computer program is executed by the processing device 501. It should be noted that the computer readable medium described in the embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In an embodiment of the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. Whereas in embodiments of the present disclosure, the computer-readable signal medium may comprise a data signal propagated in baseband or as part of a carrier wave, with computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.
The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device. The computer readable medium carries one or more computer programs which, when executed by the electronic device, cause the electronic device to:
Computer program code for carrying out operations of embodiments of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++, python and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (10)
1. The method for identifying the abnormality of the current transformer based on Hausdorff is characterized by comprising the following steps of:
Acquiring running current data of a current transformer, carrying out multi-order difference on the running current data, and screening stable current data; constructing three-phase asymmetric current components of one or more lines based on the stationary current data; according to the three-phase asymmetric current components, calculating Hausdorff distance of unbalance degree of each line and other three-phase asymmetric current components of the same bus;
Constructing characteristic parameters of line current based on the stable current data, and constructing a calculation model by using a principal component analysis method; calculating the evaluation statistic and the variation of each line according to the calculation model;
constructing a line anomaly identification model by using a first SVM algorithm according to Hausdorff distance of unbalance degree of three-phase asymmetric current components of each line and other lines under the same bus and the variation of evaluation statistics;
Constructing a phase sequence diagnosis model by using a second SVM algorithm based on a kirchhoff current law and the Hausdorff distance of an abnormal line;
And carrying out abnormality recognition on one or more current transformers of the line to be evaluated by using the line abnormality recognition model and the phase sequence diagnosis model.
2. The Hausdorff-based current transformer anomaly identification method of claim 1, wherein the constructing a line anomaly identification model by using a first SVM algorithm according to Hausdorff distance of unbalance degree of three-phase asymmetric current components of each line and other lines under the same bus and variation amount of evaluation statistic comprises:
constructing a feature vector according to the Hausdorff distance of the unbalance degree of the three-phase asymmetric current components of each line and other lines under the same bus and the variation of the evaluation statistic;
determining an objective function and a kernel function of a first SVM algorithm, and constructing a line anomaly identification model according to the objective function and the kernel function; the objective function is expressed as:
，
wherein, As the weight of the material to be weighed,A category label which indicates that the line j is normal and abnormal, and v j is a feature vector of the line j; is a threshold value; c represents a penalty coefficient and, Represents a relaxation factor; the kernel function is expressed as:， Is a gaussian radial basis function.
3. The Hausdorff-based current transformer anomaly identification method of claim 2, further comprising: and optimizing C, g parameters in the first SVM model by adopting a gold hawk optimization algorithm.
4. The Hausdorff-based current transformer anomaly identification method of claim 1, wherein the Hausdorff-based current law and the Hausdorff distance of the anomaly line, the constructing a phase sequence diagnosis model by using a second SVM algorithm comprises:
Based on kirchhoff current law, calculating Hausdorff distance between three-phase current of an abnormal line and three-phase current of other lines on the same node;
Calculating the contribution rate variation of each phase of current in the abnormal line to the evaluation statistic;
and constructing a phase sequence recognition model by adopting a second SVM based on the contribution rate variation and the Hausdorff distance.
5. The Hausdorff-based current transformer anomaly identification method of claim 4, wherein,
The contribution rate is calculated as follows:
，
wherein, The first contribution rate array cont (t) at time tA respective element, which is also the firstThe contribution rate of the station current transformer to the statistic Q (t); Denoted as time t Real-time data standardized by the mutual inductor; Is that Projection in principal component space.
6. The Hausdorff-based current transformer anomaly identification method according to any one of claims 1 to 5, wherein calculating the Hausdorff distance of the unbalance of the three-phase asymmetric current components of each line and other lines under the same bus line according to the three-phase asymmetric current components comprises:
Calculating zero sequence unbalance and negative sequence unbalance of each line based on the three-phase asymmetric current components;
And calculating Hausdorff distance between the zero sequence imbalance characteristic parameters of each line and other lines in the same period for the lines on the same bus.
7. The utility model provides a current transformer anomaly identification system based on Hausdorff which characterized in that includes:
The acquisition module is used for acquiring the running current data of the current transformer, carrying out multi-order difference on the running current data and screening out stable current data; constructing three-phase asymmetric current components of one or more lines based on the stationary current data; according to the three-phase asymmetric current components, calculating Hausdorff distance of unbalance degree of each line and other three-phase asymmetric current components of the same bus;
The calculation module is used for constructing characteristic parameters of line current based on the stable current data and constructing a calculation model by utilizing a principal component analysis method; calculating the evaluation statistic and the variation of each line according to the calculation model;
the first construction module is used for constructing a line anomaly identification model by utilizing a first SVM algorithm according to the Hausdorff distance of the unbalance degree of the three-phase asymmetric current components of each line and other lines under the same bus and the variation of the evaluation statistic;
The second construction module is used for constructing a phase sequence diagnosis model by using a second SVM algorithm based on the kirchhoff current law and the Hausdorff distance of the abnormal line;
And the identification module is used for carrying out abnormality identification on one or more current transformers of the line to be evaluated by utilizing the line abnormality identification model and the phase sequence diagnosis model.
8. The Hausdorff-based current transformer anomaly identification system of claim 7, wherein the second building block comprises:
the first calculation unit is used for calculating Hausdorff distance between the three-phase current of the abnormal line and the three-phase current of the other lines on the same node based on the kirchhoff current law;
A second calculation unit for calculating a contribution rate variation of each phase of current in the abnormal line to the evaluation statistic;
And the construction unit is used for constructing a phase sequence identification model by adopting a second SVM based on the contribution rate variation and the Hausdorff distance.
9. An electronic device, comprising: one or more processors; a storage means for storing one or more programs that, when executed by the one or more processors, cause the one or more processors to implement the Hausdorff-based current transformer anomaly identification method of any one of claims 1 to 6.
10. A computer readable medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the Hausdorff-based current transformer anomaly identification method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211203322.7A CN115469260B (en) | 2022-09-29 | 2022-09-29 | Hausdorff-based current transformer anomaly identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211203322.7A CN115469260B (en) | 2022-09-29 | 2022-09-29 | Hausdorff-based current transformer anomaly identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115469260A CN115469260A (en) | 2022-12-13 |
| CN115469260B true CN115469260B (en) | 2024-05-17 |
Family
ID=84334136
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211203322.7A Active CN115469260B (en) | 2022-09-29 | 2022-09-29 | Hausdorff-based current transformer anomaly identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115469260B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116643228A (en) * | 2023-07-27 | 2023-08-25 | 威胜集团有限公司 | Initial error estimation method for voltage transformer |
| CN117330816B (en) * | 2023-12-01 | 2024-01-26 | 南京中旭电子科技有限公司 | Monitoring data optimization method for Hall current sensor |
| CN117972460A (en) * | 2024-03-28 | 2024-05-03 | 国网福建省电力有限公司 | Operation fault discrimination method for high-voltage current transformer |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014063570A1 (en) * | 2012-10-24 | 2014-05-01 | 国家电网公司 | Method for on-line diagnosing gradually-changing fault of electronic current transformers |
| CN108572303A (en) * | 2018-04-25 | 2018-09-25 | 西南交通大学 | A kind of distribution network fault line selection method based on random matrix and Hausdorff distances |
| CN112083286A (en) * | 2020-09-07 | 2020-12-15 | 昆明理工大学 | Single-phase earth fault line selection method for wind power plant collecting line based on zero sequence current waveform similarity |
| WO2021108537A1 (en) * | 2019-11-27 | 2021-06-03 | Sentient Technology Holdings, LLC | Systems and methods for power line fault detection |
| CN115018139A (en) * | 2022-05-25 | 2022-09-06 | 国网湖南省电力有限公司 | Current transformer error state online identification method and system based on interphase characteristics |
-
2022
- 2022-09-29 CN CN202211203322.7A patent/CN115469260B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014063570A1 (en) * | 2012-10-24 | 2014-05-01 | 国家电网公司 | Method for on-line diagnosing gradually-changing fault of electronic current transformers |
| CN108572303A (en) * | 2018-04-25 | 2018-09-25 | 西南交通大学 | A kind of distribution network fault line selection method based on random matrix and Hausdorff distances |
| WO2021108537A1 (en) * | 2019-11-27 | 2021-06-03 | Sentient Technology Holdings, LLC | Systems and methods for power line fault detection |
| CN112083286A (en) * | 2020-09-07 | 2020-12-15 | 昆明理工大学 | Single-phase earth fault line selection method for wind power plant collecting line based on zero sequence current waveform similarity |
| CN115018139A (en) * | 2022-05-25 | 2022-09-06 | 国网湖南省电力有限公司 | Current transformer error state online identification method and system based on interphase characteristics |
Non-Patent Citations (2)
| Title |
|---|
| 基于Hausdorff距离的主动配电网纵联保护方案;陈星宇;;自动化技术与应用;20200325(第03期);全文 * |
| 基于幅值特征和Hausdorff距离的配电网故障定位方法;李振兴;王朋飞;王新;翁汉琍;龚旸;;电力系统自动化;20200410(第07期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115469260A (en) | 2022-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115469260B (en) | Hausdorff-based current transformer anomaly identification method and system | |
| CN109818349B (en) | Power grid robust state prediction method based on multidimensional state matrix sliding matching | |
| CN103389472B (en) | A kind of Forecasting Methodology of the cycle life of lithium ion battery based on ND-AR model | |
| CN113702895B (en) | Online quantitative evaluation method for error state of voltage transformer | |
| CN113239132B (en) | Online out-of-tolerance identification method for voltage transformer | |
| CN115469259B (en) | CT error state online quantitative evaluation method and device based on RBF neural network | |
| CN115508770B (en) | KL-NB algorithm-based electric energy meter operation state online evaluation method | |
| Liu et al. | Combined forecasting method of dissolved gases concentration and its application in condition-based maintenance | |
| CN111460374A (en) | Power distribution network D-PMU optimal configuration method considering node differences | |
| Guo et al. | A hybrid prognosis scheme for rolling bearings based on a novel health indicator and nonlinear Wiener process | |
| CN116227538B (en) | Clustering and deep learning-based low-current ground fault line selection method and equipment | |
| CN113743651A (en) | Electronic voltage transformer error state prediction method based on machine learning and SVM algorithm | |
| CN114757305A (en) | Voltage transformer insulation fault identification method and system based on ensemble learning | |
| Tang et al. | Prediction about the vessel’s heave motion under different sea states based on hybrid PSO_ARMA model | |
| CN115144807B (en) | Differential noise filtering and current-carrying grading current transformer online evaluation method and device | |
| CN115267641B (en) | Method and system for identifying error abnormity of current transformer in same-tower double-circuit power transmission line | |
| CN115616470B (en) | Method, system, equipment and medium for predicting metering error state of current transformer | |
| Jiang et al. | Two‐level area‐load modelling for OPF of power system using reinforcement learning | |
| CN114897032B (en) | Current transformer fault diagnosis method and device based on width learning | |
| CN115878963A (en) | Capacitance voltage transformer metering error prediction method, system, terminal and medium | |
| Yu et al. | Knowledge reasoning of transmission line component detection using CRITIC and TOPSIS approaches | |
| Brosinsky et al. | Machine learning and digital twins: monitoring and control for dynamic security in power systems | |
| An et al. | Transient stability margin prediction under the concept of security region of power systems based on the long short-term memory network and attention mechanism | |
| Peng et al. | Effects of sampling in monitoring power system oscillations using on-line Prony analysis | |
| Su et al. | An on-line detection method for outliers of dynamic unstable measurement data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: 430000, Unit T4, Building B12, No. 88 Guanggu Sixth Road, Donghu New Technology Development Zone, Wuhan City, Hubei Province Applicant after: Wuhan Gelanruo Intelligent Technology Co.,Ltd. Address before: Room 1803-1805, building 2-07, guanggu.core center, 303 Guanggu Avenue, Donghu New Technology Development Zone, Wuhan City, Hubei Province, 430000 Applicant before: WUHAN GLORY ROAD INTELLIGENT TECHNOLOGY Co.,Ltd. Country or region before: China |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |